Depth Map Prediction from a Single Image using a Multi-Scale Deep Network

Abstract

Predicting depth is an essential component in understanding the 3D geometry of a scene. While for stereo images local correspondence suffices for estimation, finding depth relations from a single image is less straightforward, requiring integration of both global and local information from various cues. Moreover, the task is inherently ambiguous, with a large source of uncertainty coming from the overall scale. In this paper, we present a new method that addresses this task by employing two deep network stacks: one that makes a coarse global prediction based on the entire image, and another that refines this prediction locally. We also apply a scale-invariant error to help measure depth relations rather than scale. By leveraging the raw datasets as large sources of training data, our method achieves state-of-the-art results on both NYU Depth and KITTI, and matches detailed depth boundaries without the need for superpixelation.

References in corpus (3)

• Depth Extraction from Video Using Non-parametric Sampling
• Unsupervised learning of depth and motion
• Continuous Markov Random Fields for Robust Stereo Estimation

Cited by in corpus (507)

• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• Conditional Random Fields as Recurrent Neural Networks
• Deep Visual Domain Adaptation: A Survey
• Learning Depth from Single Monocular Images Using Deep Convolutional Neural Fields
• Perceptual Losses for Real-Time Style Transfer and Super-Resolution
• Learning from Synthetic Humans
• Fully Convolutional Networks for Semantic Segmentation
• Deep Bilateral Learning for Real-Time Image Enhancement
• Multi-Task Learning for Dense Prediction Tasks: A Survey
• DeMoN: Depth and Motion Network for Learning Monocular Stereo
• Bayesian Deep Convolutional Encoder-Decoder Networks for Surrogate Modeling and Uncertainty Quantification
• AdaBins: Depth Estimation using Adaptive Bins
• FlowNet: Learning Optical Flow with Convolutional Networks
• Domain Separation Networks
• From Big to Small: Multi-Scale Local Planar Guidance for Monocular Depth Estimation
• DeepNIS: Deep Neural Network for Nonlinear Electromagnetic Inverse Scattering
• An Empirical Evaluation of Deep Learning on Highway Driving
• Perceptual Adversarial Networks for Image-to-Image Transformation
• DR2-Net: Deep Residual Reconstruction Network for Image Compressive Sensing
• SfM-Net: Learning of Structure and Motion from Video
• A Survey on Deep Learning Techniques for Stereo-based Depth Estimation
• Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue
• Monocular Depth Estimation Based On Deep Learning: An Overview
• End-to-End Learning of Geometry and Context for Deep Stereo Regression
• A2D2: Audi Autonomous Driving Dataset
• Confidence Propagation through CNNs for Guided Sparse Depth Regression
• DepthFormer: Exploiting Long-Range Correlation and Local Information for Accurate Monocular Depth Estimation
• Unsupervised Learning of Depth and Ego-Motion from Video
• InversionNet: A Real-Time and Accurate Full Waveform Inversion with CNNs and continuous CRFs
• Country-wide high-resolution vegetation height mapping with Sentinel-2
• Unsupervised Scale-consistent Depth Learning from Video
• What Makes Good Synthetic Training Data for Learning Disparity and Optical Flow Estimation?
• Learning Depth from Single Images with Deep Neural Network Embedding Focal Length
• Synthetic Depth-of-Field with a Single-Camera Mobile Phone
• High Quality Monocular Depth Estimation via Transfer Learning
• Monocular Semantic Occupancy Grid Mapping with Convolutional Variational Encoder-Decoder Networks
• Auxiliary Tasks in Multi-task Learning
• Light Field Reconstruction Using Convolutional Network on EPI and Extended Applications
• Deep Image Homography Estimation
• DISC: Deep Image Saliency Computing via Progressive Representation Learning
• DeepV2D: Video to Depth with Differentiable Structure from Motion
• AdaShare: Learning What To Share For Efficient Deep Multi-Task Learning
• Deep Shading: Convolutional Neural Networks for Screen-Space Shading
• TransCG: A Large-Scale Real-World Dataset for Transparent Object Depth Completion and a Grasping Baseline
• BA-Net: Dense Bundle Adjustment Network
• Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-Scale Convolutional Architecture
• Unsupervised Scale-consistent Depth and Ego-motion Learning from Monocular Video
• Delving into Deep Imbalanced Regression
• GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose
• Accelerating 3D Deep Learning with PyTorch3D
• Single Image Depth Estimation: An Overview
• Super-Resolution via Deep Learning
• A Survey on Deep Learning for Localization and Mapping: Towards the Age of Spatial Machine Intelligence
• Data-Free Adversarial Distillation
• Detail Preserving Depth Estimation from a Single Image Using Attention Guided Networks
• IM2HEIGHT: Height Estimation from Single Monocular Imagery via Fully Residual Convolutional-Deconvolutional Network
• Approaching the Computational Color Constancy as a Classification Problem through Deep Learning
• Single-Image Depth Perception in the Wild
• Density-aware Single Image De-raining using a Multi-stream Dense Network
• Probabilistic and Geometric Depth: Detecting Objects in Perspective
• Attribute2Image: Conditional Image Generation from Visual Attributes
• Semi-Supervised Deep Learning for Monocular Depth Map Prediction
• Monocular Depth Estimation: A Survey
• Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstruction
• Efficient piecewise training of deep structured models for semantic segmentation
• Configurable 3D Scene Synthesis and 2D Image Rendering with Per-Pixel Ground Truth using Stochastic Grammars
• Angle-Closure Detection in Anterior Segment OCT based on Multi-Level Deep Network
• Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks
• Learning 3D Shape Completion under Weak Supervision
• Pseudo-LiDAR from Visual Depth Estimation: Bridging the Gap in 3D Object Detection for Autonomous Driving
• GeoNet++: Iterative Geometric Neural Network with Edge-Aware Refinement for Joint Depth and Surface Normal Estimation
• Deep Learning-based Depth Estimation Methods from Monocular Image and Videos: A Comprehensive Survey
• Masked GANs for Unsupervised Depth and Pose Prediction with Scale Consistency
• Generalizing semi-supervised generative adversarial networks to regression using feature contrasting
• CMIR-NET : A Deep Learning Based Model For Cross-Modal Retrieval In Remote Sensing
• Unsupervised Learning of Depth, Optical Flow and Pose with Occlusion from 3D Geometry
• PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing
• Analyzing Modular CNN Architectures for Joint Depth Prediction and Semantic Segmentation
• MegaDepth: Learning Single-View Depth Prediction from Internet Photos
• Stereo CenterNet based 3D Object Detection for Autonomous Driving
• Peeking Behind Objects: Layered Depth Prediction from a Single Image
• Transformers in Self-Supervised Monocular Depth Estimation with Unknown Camera Intrinsics
• On Deep Learning Techniques to Boost Monocular Depth Estimation for Autonomous Navigation
• J-MOD${}^2$: Joint Monocular Obstacle Detection and Depth Estimation
• Unsupervised Monocular Depth Learning in Dynamic Scenes
• Self-Supervised Monocular Depth Estimation with Internal Feature Fusion
• ROCA: Robust CAD Model Retrieval and Alignment from a Single Image
• Enforcing geometric constraints of virtual normal for depth prediction
• MiniNet: An extremely lightweight convolutional neural network for real-time unsupervised monocular depth estimation
• DIODE: A Dense Indoor and Outdoor DEpth Dataset
• Pushing the Boundaries of Boundary Detection using Deep Learning
• Joint Monocular 3D Vehicle Detection and Tracking
• MGNet: Monocular Geometric Scene Understanding for Autonomous Driving
• AMNet: Deep Atrous Multiscale Stereo Disparity Estimation Networks
• Mid-Level Visual Representations Improve Generalization and Sample Efficiency for Learning Visuomotor Policies
• Semantically-Guided Representation Learning for Self-Supervised Monocular Depth
• ShaRF: Shape-conditioned Radiance Fields from a Single View
• Estimating Depth from Monocular Images as Classification Using Deep Fully Convolutional Residual Networks
• Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression
• Attention U-Net as a surrogate model for groundwater prediction
• Multi-Scale Continuous CRFs as Sequential Deep Networks for Monocular Depth Estimation
• SafetyNet: Detecting and Rejecting Adversarial Examples Robustly
• Adversarial Patch Attacks on Monocular Depth Estimation Networks
• Distortion-aware Monocular Depth Estimation for Omnidirectional Images
• Structured Attention Guided Convolutional Neural Fields for Monocular Depth Estimation
• Forget About the LiDAR: Self-Supervised Depth Estimators with MED Probability Volumes
• ReconNet: Non-Iterative Reconstruction of Images from Compressively Sensed Random Measurements
• Depth from a Single Image by Harmonizing Overcomplete Local Network Predictions
• Monocular 3D Object Detection with Sequential Feature Association and Depth Hint Augmentation
• Variational Information Distillation for Knowledge Transfer
• Pixel-wise Attentional Gating for Parsimonious Pixel Labeling
• Autonomous Driving with Deep Learning: A Survey of State-of-Art Technologies
• DiverseDepth: Affine-invariant Depth Prediction Using Diverse Data
• DeepLab2: A TensorFlow Library for Deep Labeling
• Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring
• D3VO: Deep Depth, Deep Pose and Deep Uncertainty for Monocular Visual Odometry
• Just-in-Time Reconstruction: Inpainting Sparse Maps using Single View Depth Predictors as Priors
• Exploiting the Potential of Standard Convolutional Autoencoders for Image Restoration by Evolutionary Search
• Unsupervised Monocular Depth Estimation with Left-Right Consistency
• SAFENet: Self-Supervised Monocular Depth Estimation with Semantic-Aware Feature Extraction
• Radar-Camera Sensor Fusion for Joint Object Detection and Distance Estimation in Autonomous Vehicles
• An Analysis of SVD for Deep Rotation Estimation
• Unsupervised Learning of Dense Optical Flow, Depth and Egomotion from Sparse Event Data
• Hough-CNN: Deep Learning for Segmentation of Deep Brain Regions in MRI and Ultrasound
• Monocular Depth Estimation using Multi-Scale Continuous CRFs as Sequential Deep Networks
• How hard is it to cross the room? -- Training (Recurrent) Neural Networks to steer a UAV
• DDCNet: Deep Dilated Convolutional Neural Network for Dense Prediction
• Every Pixel Counts ++: Joint Learning of Geometry and Motion with 3D Holistic Understanding
• Sparse-to-Continuous: Enhancing Monocular Depth Estimation using Occupancy Maps
• Improving 360 Monocular Depth Estimation via Non-local Dense Prediction Transformer and Joint Supervised and Self-supervised Learning
• Hierarchical Surface Prediction for 3D Object Reconstruction
• Efficient Object Localization Using Convolutional Networks
• Deep Robust Single Image Depth Estimation Neural Network Using Scene Understanding
• Holopix50k: A Large-Scale In-the-wild Stereo Image Dataset
• A Large RGB-D Dataset for Semi-supervised Monocular Depth Estimation
• A Survey on Deep Learning Architectures for Image-based Depth Reconstruction
• Self-supervised Learning with Geometric Constraints in Monocular Video: Connecting Flow, Depth, and Camera
• VarGNet: Variable Group Convolutional Neural Network for Efficient Embedded Computing
• MonoPair: Monocular 3D Object Detection Using Pairwise Spatial Relationships
• PENet: Towards Precise and Efficient Image Guided Depth Completion
• Discriminative Training of Deep Fully-connected Continuous CRF with Task-specific Loss
• Attention-based Context Aggregation Network for Monocular Depth Estimation
• Self-Supervised Joint Learning Framework of Depth Estimation via Implicit Cues
• AutoSweep: Recovering 3D Editable Objectsfrom a Single Photograph
• Depth from Monocular Images using a Semi-Parallel Deep Neural Network (SPDNN) Hybrid Architecture
• Pattern-Affinitive Propagation across Depth, Surface Normal and Semantic Segmentation
• Learning to Generate Chairs, Tables and Cars with Convolutional Networks
• Optimal Pose and Shape Estimation for Category-level 3D Object Perception
• Efficient Yet Deep Convolutional Neural Networks for Semantic Segmentation
• LEGO: Learning Edge with Geometry all at Once by Watching Videos
• Unsupervised Depth Estimation, 3D Face Rotation and Replacement
• Deep Depth Completion of a Single RGB-D Image
• Dreaming More Data: Class-dependent Distributions over Diffeomorphisms for Learned Data Augmentation
• Estimating 3D Trajectories from 2D Projections via Disjunctive Factored Four-Way Conditional Restricted Boltzmann Machines
• Deep Hybrid Real and Synthetic Training for Intrinsic Decomposition
• PLADE-Net: Towards Pixel-Level Accuracy for Self-Supervised Single-View Depth Estimation with Neural Positional Encoding and Distilled Matting Loss
• Deep Nets: What have they ever done for Vision?
• Unsupervised Learning of Monocular Depth Estimation with Bundle Adjustment, Super-Resolution and Clip Loss
• Revisiting Single Image Depth Estimation: Toward Higher Resolution Maps with Accurate Object Boundaries
• Self-supervised learning for autonomous vehicles perception: A conciliation between analytical and learning methods
• Regression Prior Networks
• Diving Deeper into Underwater Image Enhancement: A Survey
• Are Elephants Bigger than Butterflies? Reasoning about Sizes of Objects
• Deep Optics for Monocular Depth Estimation and 3D Object Detection
• DDFlow: Learning Optical Flow with Unlabeled Data Distillation
• Self-taught Object Localization with Deep Networks
• Feature-metric Loss for Self-supervised Learning of Depth and Egomotion
• Structured Knowledge Distillation for Dense Prediction
• Enhancement of Underwater Images with Statistical Model of Background Light and Optimization of Transmission Map
• Deep Multimodal Feature Analysis for Action Recognition in RGB+D Videos
• DeepLiDAR: Deep Surface Normal Guided Depth Prediction for Outdoor Scene from Sparse LiDAR Data and Single Color Image
• Advancing Self-supervised Monocular Depth Learning with Sparse LiDAR
• Sparse and Dense Data with CNNs: Depth Completion and Semantic Segmentation
• Single View Stereo Matching
• PlaneNet: Piece-wise Planar Reconstruction from a Single RGB Image
• Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild
• Robustness Meets Deep Learning: An End-to-End Hybrid Pipeline for Unsupervised Learning of Egomotion
• 3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction
• Deep Convolutional Neural Fields for Depth Estimation from a Single Image
• Adversarial Attacks on Monocular Depth Estimation
• Automatic Generation of Constrained Furniture Layouts
• Sequential Adversarial Learning for Self-Supervised Deep Visual Odometry
• Exploring Context with Deep Structured models for Semantic Segmentation
• Deep-learning-based reduced-order modeling for subsurface flow simulation
• Learning monocular depth estimation infusing traditional stereo knowledge
• Panoramic Depth Estimation via Supervised and Unsupervised Learning in Indoor Scenes
• Fast and Accurate Single-Image Depth Estimation on Mobile Devices, Mobile AI 2021 Challenge: Report
• Center3D: Center-based Monocular 3D Object Detection with Joint Depth Understanding
• Exploiting temporal consistency for real-time video depth estimation
• Analysis of Deep Networks for Monocular Depth Estimation Through Adversarial Attacks with Proposal of a Defense Method
• NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereo
• A Learning-based Framework for Hybrid Depth-from-Defocus and Stereo Matching
• Inferring Point Clouds from Single Monocular Images by Depth Intermediation
• SharinGAN: Combining Synthetic and Real Data for Unsupervised Geometry Estimation
• Visual Odometry Revisited: What Should Be Learnt?
• Two Stream Networks for Self-Supervised Ego-Motion Estimation
• RealMonoDepth: Self-Supervised Monocular Depth Estimation for General Scenes
• OCM3D: Object-Centric Monocular 3D Object Detection
• P${}^2$Net: Patch-match and Plane-regularization for Unsupervised Indoor Depth Estimation
• Learning High Dynamic Range from Outdoor Panoramas
• Moving Indoor: Unsupervised Video Depth Learning in Challenging Environments
• VisualEchoes: Spatial Image Representation Learning through Echolocation
• Deep End2End Voxel2Voxel Prediction
• A Compromise Principle in Deep Monocular Depth Estimation
• Plug-and-Play: Improve Depth Estimation via Sparse Data Propagation
• Hard-Attention for Scalable Image Classification
• Unsupervised High-Resolution Depth Learning From Videos With Dual Networks
• Geometry-Aware Symmetric Domain Adaptation for Monocular Depth Estimation
• Self-Supervised Learning for Monocular Depth Estimation from Aerial Imagery
• The Edge of Depth: Explicit Constraints between Segmentation and Depth
• Every Pixel Counts: Unsupervised Geometry Learning with Holistic 3D Motion Understanding
• Depth Information Guided Crowd Counting for Complex Crowd Scenes
• Robust Autonomous Landing of UAV in Non-Cooperative Environments based on Dynamic Time Camera-LiDAR Fusion
• Fast Calculation of Probabilistic Power Flow: A Model-based Deep Learning Approach
• Semi-Supervised Monocular Depth Estimation with Left-Right Consistency Using Deep Neural Network
• Conf-Net: Toward High-Confidence Dense 3D Point-Cloud with Error-Map Prediction
• Bidirectional Attention Network for Monocular Depth Estimation
• End-to-end depth from motion with stabilized monocular videos
• ES-Net: An Efficient Stereo Matching Network
• Consistent Video Depth Estimation
• Visualization of Convolutional Neural Networks for Monocular Depth Estimation
• Learning Depth via Leveraging Semantics: Self-supervised Monocular Depth Estimation with Both Implicit and Explicit Semantic Guidance
• Self-Supervised Learning of Depth and Ego-motion with Differentiable Bundle Adjustment
• Boosting Light-Weight Depth Estimation Via Knowledge Distillation
• Noise-Aware Unsupervised Deep Lidar-Stereo Fusion
• TOM-Net: Learning Transparent Object Matting from a Single Image
• Novel View Synthesis for Large-scale Scene using Adversarial Loss
• Self-Supervised Relative Depth Learning for Urban Scene Understanding
• Steady-state Non-Line-of-Sight Imaging
• Robust Semi-Supervised Monocular Depth Estimation with Reprojected Distances
• InverseRenderNet: Learning single image inverse rendering
• Down-Scaling with Learned Kernels in Multi-Scale Deep Neural Networks for Non-Uniform Single Image Deblurring
• Visual Depth Mapping from Monocular Images using Recurrent Convolutional Neural Networks
• Towards Better Generalization: Joint Depth-Pose Learning without PoseNet
• Deep Learning based Monocular Depth Prediction: Datasets, Methods and Applications
• Task-Aware Monocular Depth Estimation for 3D Object Detection
• Spherical View Synthesis for Self-Supervised 360 Depth Estimation
• Monocular Depth Estimation with Hierarchical Fusion of Dilated CNNs and Soft-Weighted-Sum Inference
• Self-Supervised 3D Keypoint Learning for Ego-motion Estimation
• Unsupervised Discovery of Object Landmarks as Structural Representations
• Learning to Reconstruct Texture-less Deformable Surfaces from a Single View
• Dense Depth Estimation in Monocular Endoscopy with Self-supervised Learning Methods
• DepthNet Nano: A Highly Compact Self-Normalizing Neural Network for Monocular Depth Estimation
• Learning Independent Object Motion from Unlabelled Stereoscopic Videos
• Self-supervised Learning for Single View Depth and Surface Normal Estimation
• Joint Image Filtering with Deep Convolutional Networks
• Weakly Supervised Learning of Multi-Object 3D Scene Decompositions Using Deep Shape Priors
• Learning monocular depth estimation with unsupervised trinocular assumptions
• Better Patch Stitching for Parametric Surface Reconstruction
• Moire Image Restoration using Multi Level Hyper Vision Net
• ViP-DeepLab: Learning Visual Perception with Depth-aware Video Panoptic Segmentation
• Learning to Reconstruct 3D Manhattan Wireframes from a Single Image
• Web Stereo Video Supervision for Depth Prediction from Dynamic Scenes
• Unsupervised Learning of Depth and Ego-Motion from Cylindrical Panoramic Video with Applications for Virtual Reality
• On Robust Cross-View Consistency in Self-Supervised Monocular Depth Estimation
• Towards Real-Time Monocular Depth Estimation for Robotics: A Survey
• DiPE: Deeper into Photometric Errors for Unsupervised Learning of Depth and Ego-motion from Monocular Videos
• A Lightweight Neural Network for Monocular View Generation with Occlusion Handling
• EMPNet: Neural Localisation and Mapping Using Embedded Memory Points
• Deep Shape-from-Template: Wide-Baseline, Dense and Fast Registration and Deformable Reconstruction from a Single Image
• SceneEDNet: A Deep Learning Approach for Scene Flow Estimation
• SymmNet: A Symmetric Convolutional Neural Network for Occlusion Detection
• On the Importance of Stereo for Accurate Depth Estimation: An Efficient Semi-Supervised Deep Neural Network Approach
• Beyond Photometric Loss for Self-Supervised Ego-Motion Estimation
• CAM-Convs: Camera-Aware Multi-Scale Convolutions for Single-View Depth
• Deep Depth From Focus
• Very Deep Multilingual Convolutional Neural Networks for LVCSR
• Persistent self-supervised learning principle: from stereo to monocular vision for obstacle avoidance
• Can Ground Truth Label Propagation from Video help Semantic Segmentation?
• Active Safety Envelopes using Light Curtains with Probabilistic Guarantees
• Learning Depth from Monocular Videos Using Synthetic Data: A Temporally-Consistent Domain Adaptation Approach
• Data-Driven Modeling of Geometry-Adaptive Steady Heat Transfer based on Convolutional Neural Networks: Heat Conduction
• Learning Audio-Visual Dereverberation
• Pseudo Supervised Monocular Depth Estimation with Teacher-Student Network
• Black-box Adversarial Attacks on Monocular Depth Estimation Using Evolutionary Multi-objective Optimization
• Indoor Depth Completion with Boundary Consistency and Self-Attention
• Multigrid Neural Architectures
• End-to-End Learnable Geometric Vision by Backpropagating PnP Optimization
• Learning Multi-modal Information for Robust Light Field Depth Estimation
• View Extrapolation of Human Body from a Single Image
• TW-SMNet: Deep Multitask Learning of Tele-Wide Stereo Matching
• Monocular Depth Estimation with Augmented Ordinal Depth Relationships
• Learning to Detect 3D Reflection Symmetry for Single-View Reconstruction
• Recognizing Scenes from Novel Viewpoints
• RGB-D Local Implicit Function for Depth Completion of Transparent Objects
• A multi-task convolutional neural network for mega-city analysis using very high resolution satellite imagery and geospatial data
• DeepSFM: Structure From Motion Via Deep Bundle Adjustment
• Toward Hierarchical Self-Supervised Monocular Absolute Depth Estimation for Autonomous Driving Applications
• Geo-Supervised Visual Depth Prediction
• Physical Model Guided Deep Image Deraining
• Learning to Recover 3D Scene Shape from a Single Image
• Deep Multicameral Decoding for Localizing Unoccluded Object Instances from a Single RGB Image
• Self-Supervised Generative Adversarial Network for Depth Estimation in Laparoscopic Images
• Structure-Attentioned Memory Network for Monocular Depth Estimation
• NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video
• Learning Object-specific Distance from a Monocular Image
• Unsupervised Video Depth Estimation Based on Ego-motion and Disparity Consensus
• DeepOrganNet: On-the-Fly Reconstruction and Visualization of 3D / 4D Lung Models from Single-View Projections by Deep Deformation Network
• UnDEMoN 2.0: Improved Depth and Ego Motion Estimation through Deep Image Sampling
• SLURP: Side Learning Uncertainty for Regression Problems
• SeasonDepth: Cross-Season Monocular Depth Prediction Dataset and Benchmark under Multiple Environments
• Pano Popups: Indoor 3D Reconstruction with a Plane-Aware Network
• Unsupervised object-centric video generation and decomposition in 3D
• DF-Net: Unsupervised Joint Learning of Depth and Flow using Cross-Task Consistency
• Gradient Information Guided Deraining with A Novel Network and Adversarial Training
• 3D Human Pose, Shape and Texture from Low-Resolution Images and Videos
• Automatic Generation of Dense Non-rigid Optical Flow
• Unsupervised Joint Learning of Depth, Optical Flow, Ego-motion from Video
• Monocular 3D Object Detection: An Extrinsic Parameter Free Approach
• Learning Motion Patterns in Videos
• BridgeNet: A Joint Learning Network of Depth Map Super-Resolution and Monocular Depth Estimation
• Towards Panoptic 3D Parsing for Single Image in the Wild
• SLAW: Scaled Loss Approximate Weighting for Efficient Multi-Task Learning
• Learning Depth With Very Sparse Supervision
• AutoEG: Automated Experience Grafting for Off-Policy Deep Reinforcement Learning
• Structured Depth Prediction in Challenging Monocular Video Sequences
• A Survey on Deep Geometry Learning: From a Representation Perspective
• FIS-Nets: Full-image Supervised Networks for Monocular Depth Estimation
• Unsupervised Abnormality Detection Using Heterogeneous Autonomous Systems
• CodeVIO: Visual-Inertial Odometry with Learned Optimizable Dense Depth
• Instance-wise Depth and Motion Learning from Monocular Videos
• Self-supervised Object Motion and Depth Estimation from Video
• Learning a Domain-Agnostic Visual Representation for Autonomous Driving via Contrastive Loss
• Blind Geometric Distortion Correction on Images Through Deep Learning
• Deep Photometric Stereo for Non-Lambertian Surfaces
• Virtual Normal: Enforcing Geometric Constraints for Accurate and Robust Depth Prediction
• S-BEV: Semantic Birds-Eye View Representation for Weather and Lighting Invariant 3-DoF Localization
• MINE: Towards Continuous Depth MPI with NeRF for Novel View Synthesis
• Learning to Generate Synthetic 3D Training Data through Hybrid Gradient
• Learning Joint 2D-3D Representations for Depth Completion
• Hard Pixel Mining for Depth Privileged Semantic Segmentation
• Generating and Exploiting Probabilistic Monocular Depth Estimates
• S2R-DepthNet: Learning a Generalizable Depth-specific Structural Representation
• False Negative Reduction in Video Instance Segmentation using Uncertainty Estimates
• DDCNet-Multires: Effective Receptive Field Guided Multiresolution CNN for Dense Prediction
• R4Dyn: Exploring Radar for Self-Supervised Monocular Depth Estimation of Dynamic Scenes
• Learning Multiple Dense Prediction Tasks from Partially Annotated Data
• Attentional Separation-and-Aggregation Network for Self-supervised Depth-Pose Learning in Dynamic Scenes
• Learning to See Through Obstructions with Layered Decomposition
• Learning to Navigate from Simulation via Spatial and Semantic Information Synthesis with Noise Model Embedding
• Monocular Road Planar Parallax Estimation
• Edge-Guided Occlusion Fading Reduction for a Light-Weighted Self-Supervised Monocular Depth Estimation
• LiStereo: Generate Dense Depth Maps from LIDAR and Stereo Imagery
• Robust Full-FoV Depth Estimation in Tele-wide Camera System
• What Is Around The Camera?
• Unsupervised Learning of Camera Pose with Compositional Re-estimation
• Multi-view Depth Estimation using Epipolar Spatio-Temporal Networks
• Coupled Depth Learning
• FusionMapping: Learning Depth Prediction with Monocular Images and 2D Laser Scans
• Deep feature fusion for self-supervised monocular depth prediction
• MonoIndoor: Towards Good Practice of Self-Supervised Monocular Depth Estimation for Indoor Environments
• Learning to Aggregate and Personalize 3D Face from In-the-Wild Photo Collection
• Improving Monocular Depth Estimation by Leveraging Structural Awareness and Complementary Datasets
• Depth Completion via Deep Basis Fitting
• 3D Neighborhood Convolution: Learning Depth-Aware Features for RGB-D and RGB Semantic Segmentation
• ODE-CNN: Omnidirectional Depth Extension Networks
• SIGNet: Semantic Instance Aided Unsupervised 3D Geometry Perception
• Photo-Geometric Autoencoding to Learn 3D Objects from Unlabelled Images
• Geometry-based Occlusion-Aware Unsupervised Stereo Matching for Autonomous Driving
• From Images to 3D Shape Attributes
• Inferring Distributions Over Depth from a Single Image
• Neural Autonomous Navigation with Riemannian Motion Policy
• Optical Flow Dataset Synthesis from Unpaired Images
• Learning the Depths of Moving People by Watching Frozen People
• Learning what is above and what is below: horizon approach to monocular obstacle detection
• RGB-D Image Inpainting Using Generative Adversarial Network with a Late Fusion Approach
• Symmetry-aware Depth Estimation using Deep Neural Networks
• Persistent Evidence of Local Image Properties in Generic ConvNets
• "What happens if..." Learning to Predict the Effect of Forces in Images
• LEAD: LiDAR Extender for Autonomous Driving
• Partially Supervised Multi-Task Network for Single-View Dietary Assessment
• Self-supervised Visual-LiDAR Odometry with Flip Consistency
• Deep Reflectance Maps
• Shape and Symmetry Induction for 3D Objects
• CI-Net: Contextual Information for Joint Semantic Segmentation and Depth Estimation
• Self-Supervised Learning of Depth and Motion Under Photometric Inconsistency
• SUB-Depth: Self-distillation and Uncertainty Boosting Self-supervised Monocular Depth Estimation
• Robust Consistent Video Depth Estimation
• Binary TTC: A Temporal Geofence for Autonomous Navigation
• Efficient and Scalable View Generation from a Single Image using Fully Convolutional Networks
• Single View Metrology in the Wild
• Targeted Adversarial Perturbations for Monocular Depth Prediction
• Increased-Range Unsupervised Monocular Depth Estimation
• Defocus Blur Detection via Depth Distillation
• Neural Mesh Refiner for 6-DoF Pose Estimation
• Low-viewpoint forest depth dataset for sparse rover swarms
• Monocular Depth Prediction through Continuous 3D Loss
• Domain Decluttering: Simplifying Images to Mitigate Synthetic-Real Domain Shift and Improve Depth Estimation
• Close-Proximity Underwater Terrain Mapping Using Learning-based Coarse Range Estimation
• Meshlet Priors for 3D Mesh Reconstruction
• GrabAR: Occlusion-aware Grabbing Virtual Objects in AR
• OASIS: A Large-Scale Dataset for Single Image 3D in the Wild
• Modeling 3D Surface Manifolds with a Locally Conditioned Atlas
• Adversarial View-Consistent Learning for Monocular Depth Estimation
• Robust Optimization for Deep Regression
• $S^3$Net: Semantic-Aware Self-supervised Depth Estimation with Monocular Videos and Synthetic Data
• Physical Cue based Depth-Sensing by Color Coding with Deaberration Network
• $S^3$: Learnable Sparse Signal Superdensity for Guided Depth Estimation
• Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation
• Depth Estimation from Monocular Images and Sparse radar using Deep Ordinal Regression Network
• A Neural Network for Detailed Human Depth Estimation from a Single Image
• Deep Classification Network for Monocular Depth Estimation
• GroundNet: Monocular Ground Plane Normal Estimation with Geometric Consistency
• What augmentations are sensitive to hyper-parameters and why?
• Eliminating the Blind Spot: Adapting 3D Object Detection and Monocular Depth Estimation to 360° Panoramic Imagery
• Stereo Object Matching Network
• VUNet: Dynamic Scene View Synthesis for Traversability Estimation using an RGB Camera
• Focusing on Shadows for Predicting Heightmaps from Single Remotely Sensed RGB Images with Deep Learning
• ESCT3D: Efficient and Selectively Controllable Text-Driven 3D Content Generation with Gaussian Splatting
• Self-Guided Instance-Aware Network for Depth Completion and Enhancement
• A-TVSNet: Aggregated Two-View Stereo Network for Multi-View Stereo Depth Estimation
• Monocular Outdoor Semantic Mapping with a Multi-task Network
• Self-Supervised Monocular Image Depth Learning and Confidence Estimation
• Mirror3D: Depth Refinement for Mirror Surfaces
• Extraction of Key-frames of Endoscopic Videos by using Depth Information
• Neighbor-Vote: Improving Monocular 3D Object Detection through Neighbor Distance Voting
• Single-image Tomography: 3D Volumes from 2D Cranial X-Rays
• Luminance Attentive Networks for HDR Image and Panorama Reconstruction
• Defocus Map Estimation and Deblurring from a Single Dual-Pixel Image
• Volumetric Propagation Network: Stereo-LiDAR Fusion for Long-Range Depth Estimation
• Multi-Stream Attention Learning for Monocular Vehicle Velocity and Inter-Vehicle Distance Estimation
• Dehazing Cost Volume for Deep Multi-view Stereo in Scattering Media with Airlight and Scattering Coefficient Estimation
• Cooperative Sensing in Deep RL-Based Image-to-Decision Proactive Handover for mmWave Networks
• Learning Image Matching by Simply Watching Video
• Pose2RGBD. Generating Depth and RGB images from absolute positions
• Deep Structure Learning using Feature Extraction in Trained Projection Space
• Relative Depth Estimation as a Ranking Problem
• Camera Pose Matters: Improving Depth Prediction by Mitigating Pose Distribution Bias
• DynOcc: Learning Single-View Depth from Dynamic Occlusion Cues
• Unstructured Multi-View Depth Estimation Using Mask-Based Multiplane Representation
• Video Depth Estimation by Fusing Flow-to-Depth Proposals
• Sequential Learning of Visual Tracking and Mapping Using Unsupervised Deep Neural Networks
• A Multi-Task Learning Approach for Meal Assessment
• A Front-End for Dense Monocular SLAM using a Learned Outlier Mask Prior
• How Much Depth Information can Radar Contribute to a Depth Estimation Model?
• Weakly supervised learning of indoor geometry by dual warping
• Depth Edge Guided CNNs for Sparse Depth Upsampling
• Stereo Computation for a Single Mixture Image
• Learning Monocular 3D Vehicle Detection without 3D Bounding Box Labels
• A benchmark with decomposed distribution shifts for 360 monocular depth estimation
• Weakly-Supervised Monocular Depth Estimationwith Resolution-Mismatched Data
• Physical Context and Timing Aware Sequence Generating GANs
• Veritatem Dies Aperit- Temporally Consistent Depth Prediction Enabled by a Multi-Task Geometric and Semantic Scene Understanding Approach
• Sparse2Dense: From direct sparse odometry to dense 3D reconstruction
• MonoPLFlowNet: Permutohedral Lattice FlowNet for Real-Scale 3D Scene FlowEstimation with Monocular Images
• Deep CG2Real: Synthetic-to-Real Translation via Image Disentanglement
• Learning to Parse Wireframes in Images of Man-Made Environments
• Deep Depth from Defocus: how can defocus blur improve 3D estimation using dense neural networks?
• Multi range Real-time depth inference from a monocular stabilized footage using a Fully Convolutional Neural Network
• Crowdsourced 3D Mapping: A Combined Multi-View Geometry and Self-Supervised Learning Approach
• Contrastive Feature Loss for Image Prediction
• 3D Surface Reconstruction by Pointillism
• The Aleatoric Uncertainty Estimation Using a Separate Formulation with Virtual Residuals
• Towards Comprehensive Monocular Depth Estimation: Multiple Heads Are Better Than One
• Topological Regularization for Dense Prediction
• Dilated Fully Convolutional Neural Network for Depth Estimation from a Single Image
• RGB-D Salient Object Detection Based on Discriminative Cross-modal Transfer Learning
• Depth estimation on embedded computers for robot swarms in forest
• Late or Earlier Information Fusion from Depth and Spectral Data? Large-Scale Digital Surface Model Refinement by Hybrid-cGAN
• Does it work outside this benchmark? Introducing the Rigid Depth Constructor tool, depth validation dataset construction in rigid scenes for the masses
• Indoor GeoNet: Weakly Supervised Hybrid Learning for Depth and Pose Estimation
• What Do Single-view 3D Reconstruction Networks Learn?
• Exploiting wavelength diversity for high resolution time-of-flight 3D imaging
• Predicting 3D shapes, masks, and properties of materials, liquids, and objects inside transparent containers, using the TransProteus CGI dataset
• Visual Domain Adaptation for Monocular Depth Estimation on Resource-Constrained Hardware
• Handover Management for mmWave Networks with Proactive Performance Prediction Using Camera Images and Deep Reinforcement Learning
• A Dark Flash Normal Camera
• InsertionNet -- A Scalable Solution for Insertion
• TriDepth: Triangular Patch-based Deep Depth Prediction
• Deep Two-View Structure-from-Motion Revisited
• Unsupervised Monocular Depth Estimation for Night-time Images using Adversarial Domain Feature Adaptation
• Semantic Flow-guided Motion Removal Method for Robust Mapping
• Monocular Depth Parameterizing Networks
• Unpaired Single-Image Depth Synthesis with cycle-consistent Wasserstein GANs
• PLG-IN: Pluggable Geometric Consistency Loss with Wasserstein Distance in Monocular Depth Estimation
• Neural Illumination: Lighting Prediction for Indoor Environments
• Depth Completion with RGB Prior
• On the Synergies between Machine Learning and Binocular Stereo for Depth Estimation from Images: a Survey
• Predicting Relative Depth between Objects from Semantic Features
• Learning Transparent Object Matting
• Unsupervised Learning of Depth and Deep Representation for Visual Odometry from Monocular Videos in a Metric Space
• Learning to Reconstruct and Understand Indoor Scenes from Sparse Views
• Endo-Depth-and-Motion: Reconstruction and Tracking in Endoscopic Videos using Depth Networks and Photometric Constraints
• Boundary-induced and scene-aggregated network for monocular depth prediction
• Analytical Derivatives for Differentiable Renderer: 3D Pose Estimation by Silhouette Consistency
• Outdoor inverse rendering from a single image using multiview self-supervision
• Improving Self-Supervised Single View Depth Estimation by Masking Occlusion
• Combining Events and Frames using Recurrent Asynchronous Multimodal Networks for Monocular Depth Prediction
• Synthesizing Photorealistic Images with Deep Generative Learning
• Improved Point Transformation Methods For Self-Supervised Depth Prediction
• Adversarial Structure Matching for Structured Prediction Tasks
• Position Estimation of Camera Based on Unsupervised Learning
• Indirect Domain Shift for Single Image Dehazing
• Autonomous quadrotor obstacle avoidance based on dueling double deep recurrent Q-learning with monocular vision
• Single Image Depth Prediction with Wavelet Decomposition
• Unsupervised monocular stereo matching
• 3D Vessel Reconstruction in OCT-Angiography via Depth Map Estimation
• Towards urban scenes understanding through polarization cues
• Temporally Consistent Depth Prediction with Flow-Guided Memory Units
• End-to-end Learning for Inter-Vehicle Distance and Relative Velocity Estimation in ADAS with a Monocular Camera
• Predicting Depth from Semantic Segmentation using Game Engine Dataset
• A Method for Estimating Reflectance map and Material using Deep Learning with Synthetic Dataset
• CrDoCo: Pixel-level Domain Transfer with Cross-Domain Consistency
• Domain Translation with Conditional GANs: from Depth to RGB Face-to-Face
• Soccer on Your Tabletop
• Unsupervised Learning of Depth and Depth-of-Field Effect from Natural Images with Aperture Rendering Generative Adversarial Networks
• When Autonomous Systems Meet Accuracy and Transferability through AI: A Survey